Acute aortic dissection is one of a number of conditions in medicine that carries the moniker of “the great mimicker“. But there is no doubt that it is the one with the highest jeopardy.
Acute aortic dissection (AAD) has a very high mortality, even with surgical intervention, and the time to diagnosis and then to operation is critical for patient survival. Unfortunately, because it is both rare and often atypical in its presentation, it is well recognised that delays or failure to diagnose it are common. And these can result in a poor outcome for a patient.
I remember one of the first times I went to cardiac theatre out of hours as a registrar was with a case of AAD: a 70-something man with long-standing hypertension had a typical presentation. Even with my limited experience of the anatomy of the open chest, the culprit swollen and bruised ascending aorta was clear. The surgical team told me it was going to be a long night and suggested I didn’t stay, maybe I was getting in their way in my youthful enthusiasm! Sadly, the patient didn’t survive, he at least had had the timely opportunity to.
Learning from the errors in delayed or missed diagnosis in AAD is vitally important. Tellingly it remains a common cause of medical negligence cases. So what makes the management of AAD difficult and how can we improve patient outcomes?
More knowledge and appreciation is always a good start. We do have an increasingly accurate description of the clinical paths of AAD patients. The International Registry of Acute Aortic Dissection (IRAD) particularly has made a significant contribution to our understanding over the last two decades.
Incidence & risk factors
We know AAD is the most common acute aortic syndrome, and the incidence is estimated at 5-30 per million persons per year, clearly much rarer than ACS which it is commonly misdiagnosed for, which is 4440 per million persons per year.
In the IRAD registry, the mean age was 63 years, the incidence higher in men (65%), and increases with age. Importantly, and as often in acute CV diagnoses, the prognosis is poorer in women, possibly as a result of atypical presentation and delayed diagnosis.
Hypertension is the most common risk factor (65-75%), and often poorly controlled. Other risk factors include pre-existing aortic diseases or aortic valve disease, family history of aortic diseases, history of cardiac surgery, cigarette smoking, direct blunt chest trauma and use of intravenous drugs e.g. cocaine and amphetamines.
My patient who went to theatre was quite typical and had been diagnostically straightforward, but another case I was involved in illustrates the features, though not uncommon, that can result in a delayed diagnosis and a poor outcome.
A late 30 something smoking man presents to A&E with typical myocardial ischaemic type chest pain, a +ve troponin, and inferior ischaemia on his ECG, so is put on an ACS pathway and goes to the acute medical ward on standard treatment. He is broadly well with no pulse deficit but is hypertensive. He has no past history and no family history of early CV disease or early CV death. He does have an ejection systolic murmur. His CXR is normal. He is accepted for angiography by cardiology but doesn’t have a TTE as there is no routine echo. service over the weekend. On Monday morning he is found to have a bicuspid aortic valve, with no AR, with a dilated aortic root and a clear type A dissection. After a CT he goes straight to theatre, but tragically he had significant bleeding complications, almost certainly influenced by the ACS treatment, and he didn’t live.
Diagnostic confusion with ACS is not uncommon in AAD. We recognise about 20% of patients with type A dissection have ECG evidence of acute ischaemia, in fact, it’s not that unusual for patients to come through a PPCI pathway. As ACS is so much more common, an alternative diagnosis such as AAD may not be actively explored by those initially assessing.
Also, the presenting chest pain is usually anterior and although often described as sharp, it is more often not described as tearing, or intra-scapular back pain as classically taught. It is almost always very abrupt in onset and of severe intensity. Less often are there associated features to the pain, including syncope, signs of heart failure or a CVE symptoms.
A CXR doesn’t show a widened mediastinum in one-third of cases, and no abnormality in 10%.
And, many patients still appear quite well in the throes of an AAD. Interestingly, “walk-in” patients to A&E are reported to be less likely to be correctly diagnosed.
As many patients present to A&E and medical receiving, educating the clinicians there of the clinical subtleties of AAD must help in reducing missed or delayed diagnosis. The 2010 AHA guidelines have a simple risk assessment tool and diagnostic flow chart which should help.
Lowering the index of suspicion reasonably leads to a lowering of the threshold for diagnostic imaging. But does the access to imaging influence decision making?
Access to echo can be limited for many, especially out of office hours. And although the limitation of TTE is recognised, it will often pick-up both the diagnosis and important associated findings, as seen in the second case I’ve described.
The role of “triple-rule out” scans has not found it’s place, influenced by a number of issues, including concerns over increased radiation dose, the mixed evidence-base with in general, no clear shortening of hospital stay and resource use, and hence lack of cost-effectiveness. Will this change with advances in CT scanners and/or better patient selection? Maybe.
Right place, right time
Finally, sometimes even when a patient’s diagnosis is made, getting them to the right place for optimal further management can be challenging.
Whilst doing intensive care I was involved in the care of a patient with Marfan’s who presented with typical features of aortic dissection, although there had been some initial delay in moving forward as he already had had his ascending aorta prophylactically repaired. He had presented with back pain, and his dissection started within the arch and in a type B pattern. Recognising the surgical complexity the closet surgical centre advised getting in touch and transferring to a quaternary one. It is accepted that higher volume surgical centres are essential for good patients outcomes, but by nature there are fewer places with that, meaning that travel with an acutely unwell patient, with a high potential for instability, is then necessary. After much planning, the patient travelled the length of the country to get to Oxford. I was pleased I wasn’t experienced enough to do the transfer!
In this particular case, the role of hybrid interventions with endovascular stenting of the thoracic aorta is clearly an exciting and challenging area. But this will remain the preserve of the few centres who can get the volume to build up the necessary expertise.
Key for the transferring centre is having timely co-ordination between multiple services, including cardiology, radiology, and the ITU transfer team as well as the necessary logistical support to respond promptly and on all days at all hours.
It is clear the devastating effects of AAD on patients and their family members, and we can get better. Although rare, and at times dangerously mimicking other more common diagnoses, being more clinically suspicious, having good access to diagnostic imaging, and a strong framework for timely transfers to appropriate surgical centres, are ways we can hopefully improve patient outcomes.